remove sections of XML document with Linq

Question

How would i using Linq remove all section where their element contains parameter with {} ? In my example i want to remove section with {SecName1}

Source document:

<ReceiptLayoutMaintenanceRequest>
  <ReceiptLayoutName>Test Layout1</ReceiptLayoutName> 
  <ActionName>Add</ActionName> 
 <ReceiptLayoutForMaintenance>
  <Name>Test Layout1</Name> 
  <Description>ReciptDesc</Description> 
  <PrinterName>Emulator - Receipt</PrinterName> 
  <ReceiptLayout>
    <Name>AAA</Name> 
    <Description>$</Description> 
    <TemplateName>DefaultTemplate</TemplateName> 
    <LayoutParameters /> 
  </ReceiptLayout>
  <ReceiptLayout>
    <Name>{SecName1}</Name> 
    <Description>$</Description> 
    <TemplateName>DefaultTemplate</TemplateName> 
    <LayoutParameters /> 
  </ReceiptLayout>
 </ReceiptLayoutForMaintenance>
</ReceiptLayoutMaintenanceRequest>

Wanted output

<ReceiptLayoutMaintenanceRequest>
  <ReceiptLayoutName>Test Layout1</ReceiptLayoutName> 
  <ActionName>Add</ActionName> 
 <ReceiptLayoutForMaintenance>
  <Name>AAA</Name> 
  <Description>ReciptDesc</Description> 
  <PrinterName>Emulator - Receipt</PrinterName> 
  <ReceiptLayout>
    <Name>AAA</Name> 
    <Description>$</Description> 
    <TemplateName>DefaultTemplate</TemplateName> 
    <LayoutParameters /> 
  </ReceiptLayout>
 </ReceiptLayoutForMaintenance>

thanks

BrokenGlass · Answer 1 · 2011-10-16T16:36:44.873

2

This removes any ReceiptLayout node which has a child Name that starts and ends with brackets and produces your desired output:

XDocument doc = XDocument.Load(@"test.xml"); //load xml
var nodesToRemove = doc.Descendants("ReceiptLayout")
                       .Where(x => x.Element("Name").Value.StartsWith("{") 
                                && x.Element("Name").Value.EndsWith("}"))
                       .ToList();

foreach (var node in nodesToRemove)
    node.Remove();

This can be shortened into one Linq statement, personally I prefer to keep Linq query and modification (removal) separate though:

doc.Descendants("ReceiptLayout")
   .Where(x => x.Element("Name").Value.StartsWith("{") 
            && x.Element("Name").Value.EndsWith("}"))
   .Remove();

edited Oct 16 '11 at 16:36

answered Oct 16 '11 at 16:01

BrokenGlass

158,293
28
286
335

The first code sample gives me `NullReferenceException` in the foreach. Probably it doesn't like that you touch his collection while he is working on it. The second one works. – xanatos Oct 16 '11 at 16:31
Adding a ToList in first query should fix it. – Muhammad Hasan Khan Oct 16 '11 at 16:35
@xanatos: Good catch, I don't know how I overlooked this - added a `ToList()` in the first example, which solves the problem – BrokenGlass Oct 16 '11 at 16:37

Muhammad Hasan Khan · Answer 2 · 2011-10-16T16:35:25.077

1

var doc =  XDocument.Parse(xml);
doc.Descendants()
   .Where(n => !n.HasElements  && Regex.IsMatch(n.Value, "[{].*?[}]"))
   .Select(n=>n.Parent) // because you want to remove the section not the node
   .Remove();
xml = doc.ToString();

edited Oct 16 '11 at 16:35

answered Oct 16 '11 at 16:03

Muhammad Hasan Khan

34,648
16
88
131

Sorry but 2 answers doesnt help. Still i can't remove this section( – user681055 Oct 16 '11 at 16:32
@user681055 there was glitch in regex. Try now. – Muhammad Hasan Khan Oct 16 '11 at 16:35

xanatos · Answer 3 · 2011-10-16T16:42:32.247

A variant of BrokenGlass code using the let keyword

var doc = XDocument.Load(@"test.xml");

var list = from p in doc.Descendants("ReceiptLayout")
           let q = p.Element("Name")
           let r = q != null ? q.Value : string.Empty
           where r.StartsWith("{") && r.EndsWith("}")
           select p;

list.Remove();

This is "premature optimization" :-) I "cache" the p.Element("Name").Value. Ah... And I check if really there is a Name Element so that everything doesn't crash if there isn't one :-)

remove sections of XML document with Linq

3 Answers3